This book project is the outcome of an international conference on ‘Equity in higher education: evidence, policy and practice’ (Porto, 5–6 December 2019), jointly organised by CIPES and A3ES and funded by EDULOG (Foundation Belmiro de Azevedo). The conference focused on social inequalities in participation in higher education (regarding access and success).

With the emergence of the knowledge society many individuals now require theoretical knowledge to perform their work, and a well-educated workforce has become the major resource of the post-industrial societies (Amaral, 2018). This has prompted countries to expand their higher education systems and ensure that the population has increasing higher levels of education. Quality assurance systems have been implemented to guarantee the quality of education provision and concerns about access equity have been triggered as it was recognised that significant layers of the population did not reach the expected education levels. The present pandemic situation, by taking the lives of a disproportionate number of people from deprived strata of the population has made visible the unfair living conditions of part of the population and may increase equity concerns beyond those related to access to education.

Widening access to higher education became an issue of political concern following the end of the Second World War. It aimed both at improving social justice by granting conditions of access to all social groups and improving economic performance, because a more educated population would contribute to the economic competitiveness of countries. In his chapter Per Olaf Aamodt quotes Erikson and Jonsson (1996) to present additional reasons for implementing policies aiming at eliminating inequalities in educational enrolments, which are seen as being socially ineffective and unfair as a person’s life chances should not be dependent of the status of the families they were born into, while creating social discord from reinforced class differences and inducing lack of representativeness if all people with higher education, holding leading positions in society, were recruited from a narrow social elite (Erikson & Jonsson, 1996).

There was a generalised conviction that social inequalities could be reduced through the massification of higher education. However, although some progress has been made in terms of increased participation, the goal of equal treatment of valid student applicants, independently of their socioeconomic backgrounds (e.g. fairness), has persistently remained out of reach. This book aims to analyse the causes that make equity of access (in the sense of fairness) such a difficult objective to fulfil. There was a deliberate choice to address primarily the socioeconomic dimension, being aware that the socioeconomic background is closely related to ethnicity and race, especially since these latter condition the former to a great extent. Data on these variables is not available in all the studied national contexts (e.g. Portugal), but race and ethnicity are discussed in the case of countries where this is possible (U.S. and Brazil).

An initial chapter presents the theoretical background which relates the difficulties in attaining fairness to the fact that higher education is a positional good, and positional goods have a tendency to be monopolised by social groups from privileged backgrounds (Marginson, 1998, 2011). The chapter also presents two hypotheses (Maximally Maintained Inequality and Effectively Maintained Inequality) which have been used to describe inequality patterns. This initial chapter is followed by several chapters presenting national cases with a description of policies aiming at increasing fairness and the reasons why these policies were not successful. The chapters include contrasting examples. Two of the countries (U.S. and Brazil) have implemented affirmative action policies. However, although they both have a federal organisation, in the U.S. the federal government does not have an important role in education as educational policies come under the purview of each state, while in Brazil the federal government has control over higher education and even finances a network of federal universities. In the U.S. there are no regulations on tuition fees and each institution can decide freely on the recruitment of students. On the contrary, in Brazil the federal government centralises the regulation of the higher education system. Another country (England) was chosen because it has aimed at improving social justice within a context of neoliberal policies. The Nordic countries offer a contrasting example, as they are based on the social democratic model characterised by redistribution policies and universalistic contribution mechanisms, being very far from the neoliberal worldview. At last, a Southern European country (Portugal) completes the chosen sample. A final chapter presents the conclusions.

It is also important to stress that many countries have developed cost-sharing policies transferring to students and their parents at least part of the education costs. To make this change more tolerable governments have also created loan systems which can be of two types. In the case of mortgage-style loans (e.g. U.S.) students start repaying the debt once they complete their studies and instalments are calculated for a specified repayment period based on the total amount of the loan plus interest. Mortgage-style loans are rather risky for students who may default if they do not complete their studies or if their earned income is excessively low. Income-contingent loans (e.g. England) offer more protection to students as default, in principle, is not possible. Graduates only start repaying their loans when their earnings go above a fixed threshold, when they start paying a percentage of the income about that threshold. Payments cease when the debt is paid or after a period of 25 to 30 years, when any outstanding debt is written off. Payments are directly deducted from salaries via the tax system and students will only escape payments if they migrate to another country. This is apparently a fair system and some kind of group insurance. If one gets a very good job once completing the studies, the debt will be paid in a short time; otherwise, payments will be adjusted to earnings and may even be null if one is unemployed or poorly paid.

In Chap. 2 Alberto Amaral presents a definition of equity which is diverse from the notion of equality, and refers several supranational organisations that have proposed declarations furthering access equity. This is followed by an analysis of the components of equity which comprise fairness and inclusion. However, equity is not restricted to access; it also needs to include a component of success, as the socioeconomic background of parents is one of the strongest predictors of students’ academic achievement and attainment (Reardon, 2011). Indeed, supranational organisations recommend the implementation of policies promoting the success of students, as access without success is rather meaningless (IAU, 2008).

Massification of higher education systems was accompanied by their diversification in order to attend to the very diverse needs and aspirations of a much more heterogeneous student population. In most countries, diversification was achieved by implementing new types of institutions different from research universities, many offering shorter degrees. Diversification has created a new, subtler form of inequity, as students from deprived backgrounds have a tendency to concentrate in these new lower value opportunities (Koucký et al., 2010; Shavit et al., 2007).

The main argument of the chapter is focused on the idea that education is a positional good in the sense that it provides students with a competitive advantage when looking for employment, social standing and status (Marginson, 1998). The problem is that positional goods tend to be monopolised by social groups from privileged backgrounds, which contributes to the persistence of inequities in higher education. To interpret inequality patterns among different cohorts of students related to diverse socioeconomic backgrounds Amaral (this book) resorts to the hypotheses of Maximally Maintained Inequality (MMI) (Raftery & Hout, 1993) and Effectively Maintained Inequality (EMI) (Lucas, 2001). While a higher education system expands and before it becomes saturated, privileged economic groups use their advantages to ensure that their children can secure a place (MMI). Once the system comes close to saturation, the game changes, and as almost everyone can enter higher education, the socioeconomically advantaged look for qualitative differences, and try to secure places in the best institutions and the best study programmes, which is consistent with the idea of education as positional good. MMI and EMI describe inequality patterns but do not explain them. Amaral quotes Raftery and Hout (1993) who used rational-choice theories for an explanation. In Chap. 9, Per Olaf Aamodt discusses a number of theories which aim at explaining why social background influences inequality in education: the value theory, the cultural theory (Bourdieu & Passeron, 1977) and the social position theory (Boudon, 1974). Amaral refers to the argument of the OECD that equity in tertiary education is affected by inequities in preceding levels of education and concludes with a discussion on a number of factors which can favour or hinder equity such as early tracking systems, the selective role of mathematics and national traditions which rule access to tertiary education.

In Chap. 3 David Dill presents a critical analysis of the U.S. case which is extremely relevant being the first country to promote massification of its higher education system. David Dill questions the effectiveness of current American policies governing access to higher education, namely in terms of equity of access and the promotion of higher earnings mobility. There is lack of national regulation of tuition and fees in private higher education where tuition can go to very high levels. Even in public universities, despite state-level efforts to limit tuition fees, their net value per full-time equivalent student has increased by 96% over the last 25 years (SHEEOA, 2019) while per capita state appropriations declined by 8%. And institutions are completely free in deciding which new students will be recruited; this, as David Dill argues, can have dramatic consequences in terms of equity.

In the absence of state regulation, it is expected that the market will operate. However, for a market to operate efficiently (Leslie & Johnson, 1974), one of the conditions is that purchasers have a good knowledge of the price and characteristics of the goods and services to be purchased, as well as the market conditions. David Dill (this book) quotes Nicholas Barr (2010) who argues that students applying to a higher education place are in general well-informed customers able to make rational economic choices, which contrasts with Dill’s argument that students are immature customers lacking enough information to make, discretionary choices (Dill, 1997). Indeed, lack of information has been, for many years, a recurrent problem of American higher education, and students may enrol in institutions with very low completion rates or offering diplomas with very low employment prospects.

The Commission on the Future of Higher Education appointed in 2005 by Margaret Spellings, U.S. Secretary of Education, recommended that accreditation decisions should be more based on evidence of student achievement and institutional performance, and the final reports should be made public (Commission of the Future of Higher Education, 2006). However, Margaret Spellings was defeated in Congress and no change was possible.

In the Obama administration Secretary of Education Arne Duncan (2015) also tried to change the situation. He presented as an example, citing a Wall Street Journal investigation, the case of 11 schools with six-year graduation rates below 10% which still managed to get accreditation. He complained of very high drop-out rates which left many people with a debt from loans and that some schools offered diplomas which meant little to nothing to employers. The Obama administration tried to implement a rating system in order to promote greater transparency and offer more information to students and their families. This project was met with strong opposition from higher education institutions and their leaders and was later abandoned in favour of the College Scorecard. ‘The College Scorecard (2015) [introduced by the Obama administration] provides students, families, guidance counsellors, non-profits and other key stakeholders with institutional data through an online tool which has been accessed by over 2.5 million users (Kreighbaum, 2017). Users could find average annual cost of an institution, its graduation rate, the typical salary post-graduation, and information on student debt levels’ (Ransom et al., 2018, p. 17).

The colleges’ freedom to decide admissions in the U.S. allows them to use a method known as ‘holistic review’ which has a historical origin in a strategy of elite colleges developed before the Second World War, ‘in part to deny admission to growing numbers of Jewish applicants’ (Bastedo et al., 2017, p. 1). Holistic admissions are ‘defined as evaluating prospective students in the context of the educational, personal, and financial conditions experienced by the applicant (Bastedo & Bowman, 2017; College Board, 2012; Lucido, 2014)’ (Bowman & Bastedo, 2018, p. 431). This means that decisions on admission are based not only on the numeric results of tests such as SAT (Scholastic Aptitude Test) or ACT (American College Tests); they also include a large number of qualitative data such as essays, teacher/counsellor recommendation, demonstrated interest, internships, volunteerism, class rank, interview, geography, race/ethnicity, athletic ability, leadership, personal qualities, portfolios, veteran status, first generation status, legacy, and so on.

This means that decisions on student recruitment lack transparency. Many institutions include what is known as ADLC criteria, based on athletic ability (A), a list composed by the Dean (D), legacies made by the parents of prospective students (L) and staff children (C). Dill quotes a paper by Arcidiacono et al. (2019) which analyses a court case against Harvard to show that in this university ADLC students have a much higher probability (from 86% for athletes to 33.6% for legacies) of getting a place than other students (below 8%).

Initially some institutions established racial quotas but they were declared unconstitutional by the U.S. Supreme Court. However, race/ethnicity has been used in the holistic admissions system. As stressed by David Dill, as it became recognised that a person’s life chances could be influenced by the level of education, there has been an increasing public opposition to affirmative action and a survey has shown that 73% of Americans did not favour the use of race/ethnicity in admission decisions (Pew Research Centre, 2019). Several cases were also taken to court, but so far the Supreme Court has upheld the principle that ‘a diverse student body is essential to the educational objectives of colleges and universities’ (AAUP, 2020), but the University must ‘ensure that each applicant is evaluated as an individual and not in a way that makes an applicant’s race or ethnicity the defining feature of his or her application’ (Fisher vs. University of Texas, 2013). In a recent decision on a case against the University of Harvard the judge although upholding the integrity of holistic review recommended that ‘statistics should be used as a check on the process and as a way to recognize when implicit bias might be affecting outcomes’ (Alvero et al., 2020, p. 202). Recent cases of corruption, namely bribery of coaches to declare potential students as athletes has again raised the question of the fairness of access practices (Jaschik, 2019a, 2019b).

Finally, David Dill refers to the U.S. loan system which uses a mortgage approach, which has created serious difficulties for many young people who have defaulted on their loans and which has led to a huge accumulated debt of over one trillion dollars. At the same time, the movement towards merit-based financial loans has created additional inequalities, since in general students from affluent backgrounds are more likely to be contemplated, as family background has influence over the performance of students.

In Chap. 4 Bertolin and McCowan present the Brazilian case which also includes a component of affirmative action policies. Brazil is one of the most unequal countries in terms of income distribution (the 10th worst country as measured by the Gini coefficient [UNDP, 2016]) and one of the last moving to mass higher education. Despite an impressive increase in enrolments over the last three decades, from 1.5 million to more than 8 million, the percentage of graduates aged 25 to 34 in Brazil was only 18%, as compared against an OECD average of 43% (OECD, 2018).

One of the characteristics of Brazilian education consists in the low quality of public secondary schooling when compared against its private counterpart, while public higher education is recognised has having superior quality when compared with private higher education. This situation has strong reflection in enrolments in higher education. Students from deprived backgrounds enrol in public secondary schools as they cannot afford to pay to attend a private school, and they are at a disadvantage when competing with students from private schools for a place in a public higher education institution, where there is no tuition. In most cases they will find a place in a lower-quality higher education institution where they have to pay significant tuition fees.

During the period 1990–2002 enrolments almost doubled, but in the absence of affirmative action policies this expansion benefited mainly students from the middle and upper-middle classes while the participation of students from deprived backgrounds (bottom half of the income distribution) fell from 8.6% to 7.5%. In 2002, no student belonged to the 20% poorest of the population and only 4% belonged to the 40% poorest (Bertolin & McCowan, this book). These data are consistent with the MMI hypothesis. Later, the government of President Lula da Silva introduced affirmative action policies for the public sector. Act 12.711/2012, the Quotas Law, established a 50% quota for public school students, divided into sub-quotas for low-income students and black, mixed-race and indigenous students. A recent study of the Brazilian Institute for Geography and Statistics (IBGE, 2019) reported that for the first time the number of black and mixed-race students in public higher education institutions surpassed the number of white students.

For the private sector institutions, Act 11.096/2005 created a financing programme (Prouni, University for all) providing non-refundable full and partial (50%) scholarships for enrolments at private colleges and universities. This programme has been a bounty for the private sector, including for-profits and the sector grew from 58% of total enrolments in 1995 to 75% in 2018 (INEP, 2019). However, much of this expansion resulted from an explosive increase of distance education (DE) in the private sector, from 2% of total enrolments in private higher education in 2005 to some 30% ten years later. In 2018 the number of new DE places in private higher education was already higher than the number of places in classroom-based study programmes. However, while 30.7% of young whites were enrolled in higher education institutions in 2018, only 15.1% of black youngsters and 16.3% of mixed-race youngsters were enrolled (Todos pela Educação, 2019). In terms of secondary school graduates, only 33% of black and mixed-race students were enrolled in higher education institutions to be compared with 52% of white students (IBGE, 2019). However, students from deprived backgrounds tend to concentrate in inferior quality study programmes offered by private institutions in the distance education mode, of lower social value but at the same time less expensive (tuition fees are about 1/3 of classroom-based programmes).

It is undeniable that the share of students from lower socioeconomic levels among enrolees and graduates has increased significantly even in the high prestige federal sector (Bertolin & McCowan, this book), even if they are more likely to finish their degrees in lower-quality institutions and in courses that are less socially and economically valued (Bertolin & McCowan, 2020). On the contrary, higher valued study programmes such as Medicine have lower enrolments for non-white students, students from low-income families, graduates from public secondary schools, and students whose mothers had little education (Bertolin & McGowan, this book), which is consistent with the Effectively Maintained Inequality (EMI) hypothesis.

Bertolin and McGowan conclude that there are increasing participation rates for students admitted through affirmative action policies (Quotas Law) or public funding (e.g. Prouni—a scholarship programme and Fies—a public loan scheme), which democratise access. However, there are still remnants from an elitist system maintaining an inappropriate level of inequity.

The next two chapters deal with the English higher education system. Claire Callender presents a critical analysis of the English policies for funding higher education and its consequences, and Liz Thomas reflects on English policies aiming at improving access and completion by under-represented and vulnerable groups.

In her chapter, Claire Callender considers that reforms in England were designed under the influence of neoliberal policies and the idea of the marketisation of higher education. Since the 1990s England has introduced cost-sharing policies aiming at expanding the higher education system and increasing participation rates. Those policies have progressively transferred the costs of higher education to students and/or their families by introducing tuition fees and replacing grants with loans. It should be noted that loans can be mortgage-style or income-contingent. Chapman argues that students who default ‘face damage to their credit reputation and thus eligibility for other loans, such as for a home mortgage’ (Chapman, 2006, p. 82). The U.S., where loans are mortgage-type, is confronted with a huge student debt of over one trillion US dollars.

In England loans are income-contingent and students start paying 9% of their earnings above the threshold level of £25,000 per year and any outstanding debt is written off after 30 years. Tuition fees were introduced in 1998 by Tony Blair’s Labour government, with a value of £1000. That value was increased to £3000 in 2006/07 following the passing of the 2004 Higher Education Act by Tony Blair’s Labour Government. In 2012/13, following the recommendations of the Browne report (2010), the Cameron-Clegg’s coalition government of the Conservative Party with the Liberal Party increased fees to £9000. In 2017/18 fees were increased to their present value of £9250. Fees for part-time students with a maximum value of £6750 were introduced in 2012/13 and were increased to £6935 in 2017/18. The government hoped that higher education institutions would compete on price by charging different tuition fees but these expectations were completely frustrated as virtually all institutions charged the maximum value, one of the reasons being that charging lower fees could be seen as an indication of lower quality. And any occasional loss in enrolments would be more than compensated by the higher prices.

The government also hoped that as higher education institutions were made more financially self-sufficient, the system could expand and the number or available places would increase. This was even made easier as the cap government placed on the number of students, which universities could recruit has been progressively lifted, being completely abolished in 2015/16. Unfortunately, this prediction failed. The total number of students fell from a maximum of 851,590 in 2008/10 to 666,210 in 2012/13 and then remained almost stable until 2018/19 when enrolments totalled 674,840. However, the behaviour of full-time students was different from that of part-timers. Enrolments of full-time students had reached a maximum of 521,605 by 2011/12, then dropped to 466,270 in 2012/13 just after the large increase in fees to £9000 and then recovered slowly to 546,305 in 2018/19. On the contrary, enrolments of part-time students reached a maximum of 344,775 in 2008/09, decreasing to 199,940 in 2012/13, and then continued to drop to a value of 128,535 in 2018/19.

An additional objective of the government, which also remained unfulfilled, was to increase the relative participation rates of students from under-represented groups, another objective which remained unfulfilled. Claire Callender (this book) presents data demonstrating that the gap in participation rates between students from the poorest backgrounds and their wealthier peers actually increased from 18.5% to 21.6% from 2010/11 to 2017/18.

Claire Callender explains the sharp decline in the enrolments of part-time students by the fact that almost half of them did not qualify for loans (53% in 2015, Callender & Thompson, 2018). Moreover, many of those who qualified for a loan would not take it (41% in 2012, ibid.), one of the reasons being that they were in general employed and the eventual increase in salary once they graduated might not compensate the investment made. At last, debt aversion also plays a role, as ‘students from lower socioeconomic backgrounds are more sensitive to financial incentives … and are less willing to borrow than students from higher socioeconomic backgrounds’ (Vossensteyn & De Jong, 2006, p. 239). Indeed, even if Barr considered debt aversion irrational in the case of income-contingent loans (Barr, 2010), Callender and Mason (2017) found that debt aversion affects higher education applications from young working-class students.

The higher education reform which included a very substantial increase in student fees was publicly presented as a ‘political “compromise” of widening access to higher education to promote social justice, and improving the economic capacity of individuals and the nation’ (Liz Thomas, this book). Liz Thomas argues that the English higher education system has exhibited both characteristics of Maximally Maintained Inequality (MMI) and of Effectively Maintained Inequality (EMI). The Robbins Report (Committee on Higher Education, 1963) recommended a major expansion of the higher education system in order to provide places for ‘all who were qualified for them by ability and attainment’, independently of their family background. However, data from the 1997 Dearing Report showed that the extra places resulting from the expansion of the higher education system in the 1980s and 1990s were mainly taken up by middle class students, namely women, leaving students from lower socioeconomic groups and ethnic minority groups significantly under-represented, especially in the traditional universities (Thomas, this book), which is consistent with the MMI hypothesis (Thomas, 2001). The expansion of the number of graduates undermined the positional value of higher education and students from privileged classes started to look for differentiation by enrolling in elite institutions and pursuing graduate studies (Thomas, this book) which is consistent with the EMI hypothesis (Lucas, 2001).

Liz Thomas dedicates most of her chapter to discussing policies that were implemented in England to promote retention and completion rates. It is well-known that students from lower economic status backgrounds have higher odds for not completing their study programmes (HEFCE, 2013) (see also Bertolin & McCowan, this book). Given the very high level of tuition fees, not completing a degree would leave students with a substantial debt to pay and under an income-contingent loan scheme this debt would be transferred to the State. Indeed, access without success is a deceitful offer (Tinto, 2008) and England introduced a number of policies to promote equality of outcomes, or equity, through inclusion (Marginson, 2011).

When the value of tuition fees increased to £3000 in 2006/07 the government also installed the Office for Fair Access (OFFA), an independent, non-departmental public body with the mission of promoting and safeguarding fair access to higher education for lower income and other under-represented groups. This body was replaced with the Office for Students (OfS) from 1 April 2018. Initially the emphasis of OFFA was on the access of students from lower economic status backgrounds and institutions proposing to charge higher tuition fees had to agree with OFFA on an ‘access agreement’ explaining how they intended to safeguard and promote fair access to higher education. In 2018 access agreements were replaced with ‘Access and Participation Plans’ which detail how institutions will improve equality of opportunity for under-represented groups to access, succeed in and progress from higher education. These plans need previous approval by the Office for Students.

To promote the quality of higher education provision the government in England introduced the Teaching Excellence and Student Outcomes Framework (TEF). TEF is an award given to institutions that go beyond the national quality standards, ensuring excellent outcomes for their students in terms of graduate-level employment or further study. The awards can be gold, silver, bronze or conditional and they contain information on student satisfaction, employment outcomes and the number of students who continue their studies from one year to the next, helping prospective students to choose the institution where they want to enrol.

Liz Thomas argues that although these policies have been less successful in terms of increasing the participation of students from non-traditional groups, especially in pre-1992 elite universities, they have contributed to reduce the gap in retention and completion rates between students from diverse socioeconomic backgrounds and different ethnicities.

The next two chapters present the Portuguese case and show that developments in Portugal were consistent with the MMI and EMI hypotheses. Following the 1974 revolution, the Portuguese higher education system went through a process of accelerated expansion, with gross participation rates jumping from around 7% in 1974 to more than 50% at present. It is possible to argue that there are two distinct profiles in Portugal; the population over 35 years old which has in general a low education level and a younger sector, up to 35 years of age, with an expressive percentage of higher education graduates.

In secondary education there are two tracks: an academically oriented traditional one and a vocational one. Higher education is diversified, with universities and polytechnics (in the meaning of the former UK polytechnics), public or private. Higher education degrees are compatible with the Bologna structure and non-degree awarding shorter study cycles (TESP) were recently introduced for students willing to make an early entrance into the labour market.

The OECD (2008) presented a number of factors influencing equity in the access to higher education, including inequalities present in earlier phases of the education system (Marcenaro-Gutierrez et al., 2007; Wömann & Schütz, 2006), family socioeconomic background, early tracking, geography, articulation between secondary and tertiary education, access system. In their chapter, Baptista, Sin and Tavares discuss how these factors contribute to social inequity in Portugal, still present despite the massive increase in enrolments.

Baptista, Sin and Tavares (this book) argue that socioeconomic status is the most important aspect that directly or indirectly conditions transition to higher education in Portugal, and thus is the primary source of inequality in participation, with students whose parents have higher educational capital (in general also associated with higher income) being over represented in higher education. The socioeconomic filter in upper secondary education has origins in students’ trajectories and academic performance in basic and lower secondary education. Research results for Australia, England, U.S. and Canada show that the attendance of private schools appears to favour entry into higher education (Chesters & Watson, 2013; Engberg & Wolniak, 2010; Frempong et al., 2012; Mangan et al., 2010) and the same is apparently true for Portugal. To make things worse, it was found that there are some schools (mainly private) which systematically inflate grades to give an extra advantage to their students (the Inspectorate for Education and Science has just announced that those schools found guilty of grade inflation will be punished). As students from deprived backgrounds cannot afford to pay for a place in a private school this is an additional factor of inequality of access to higher education.

The existence of two main tracks, one more academic and one more vocational, plays an important role in maintaining inequality as the vocational track not only prepares students for an early entrance into the labour market, but also places some barriers to transition to higher education in Portugal (the government very recently passed legislation to soften these barriers). In 2017/18, 80% of graduates from the academic track were enrolled in higher education one year later, almost all in degree awarding programmes. In contrast, only 18% of those who graduated from the vocational track (disproportionately attended by disadvantaged students) were enrolled in higher education after one year, two thirds of which in TESP programmes.

Using social support (which is means tested) as a proxy for socioeconomic status, Baptista, Sin and Tavares demonstrate that the stronger the level of support, the lower the proportion of secondary school graduates that pursue higher education studies. Data also show that, on average, students who receive higher levels of social support tend to be given lower internal grades, when compared to their higher status peers that have similar scores in the national exams.

There are also significant regional differences in the enrolment rates in higher education among upper secondary graduates from different Portuguese regions, especially in the case of graduates of secondary vocational programmes, which is compatible with the results presented by the OECD (2008). This is the result of places in higher education programmes and institutions not being equally accessible to local students throughout the country, which places an extra burden on students from deprived backgrounds due to the costs of displacement from their household.

In Chap. 8, Sá, Tavares and Sin present further analyses of the Portuguese higher education system aiming at understanding why there are still persistent inequalities despite the expansion of the system since the 1974 revolution. In the first part of their chapter the authors examine the influence of the students’ socioeconomic and cultural backgrounds on their choice of institution and study programme. The expansion of the system was achieved through diversification—a dual system of universities and polytechnics—and privatisation—a parallel system of public and private institutions. This expansion has improved the chances of students from lower socioeconomic backgrounds to enter higher education (inclusion), which is consistent with the MMI hypothesis. However, diversification has resulted in social stratification, with universities having more prestige than polytechnics.

Empirical results show that students from privileged backgrounds prefer universities while students from lower socioeconomic backgrounds tend to concentrate in polytechnics. Sá, Tavares and Sin (this book) use the parents’ qualification (holding or not a higher education degree) to show that students with parents holding higher education degrees are overrepresented in universities relative to their presence in the Portuguese society, while students from families with lower cultural capital are under-represented. On the other hand, polytechnic institutions enrol a more diversified student body, being more representative of the composition of the student population in Portugal, which is consistent with the EMI hypothesis. On average families with higher cultural background are also expected to be those with higher socioeconomic capital. In Portugal, scholarships are means tested which makes the percentage of scholarships a good proxy for the socioeconomic level of families. Sá, Tavares and Sin also show that the percentage of students with scholarships is higher in polytechnic institutions than in universities, which is consistent with the results obtained for cultural capital.

The persistence of inequalities is also visible in the choice of study programmes, as students from families with higher cultural or socioeconomic capital have a much higher enrolment rate in the most prestigious study programmes, when compared with students from less privileged backgrounds. Sa et al. illustrate these findings with several examples of study programmes from similar scientific areas, but from different higher education sub-systems [e.g. Medicine (university) against Nursing (polytechnic); Pharmaceutical Sciences (university) against Pharmacy (polytechnic); Design (university) against Design (polytechnic)].

In the second part of their chapter, Sá, Tavares and Sin use two datasets, one containing data on individual candidates to public higher education and another one containing programme/institution level data. The first dataset contains data on individual candidates to public higher education for the period 2012 to 2018 (almost 330,000 individuals). The second dataset refers to the academic year 2017/18 and contains information on the proportions of candidates to income-based scholarship and scholarship holders, as well as on the parental educational levels (mother and father, separately). There is also information on the minimum admission GPA (grade point average) and admission exams, for each pair programme/institution. Despite the fact that the number of places in public higher education institutions is close to the number of candidates, every year a number of students is unable to find a place (11.6% in the period under analysis). Sa et al. analyse the characteristics of the left-out students to conclude that candidates to places in Social Sciences, Business and Law have more difficulty in getting a place; this is a consequence of the high number of candidates due to the fact that the mathematics exam is not a compulsory admission criterion for most study programmes. Unsuccessful candidates are also over-represented in programmes such as Medicine, but this is due to the very selective and demanding nature of the admission criteria. There are also significant regional differences, with unsuccessful candidates being over-represented in Porto and Lisbon, which is due not only to very high population density but also to the fact that institutions in these urban areas attract a large number of candidates from other regions.

Next, Sá, Tavares and Sin examine inequalities within the public higher education system using four different models. The first two models estimate inequalities among placed students. Consistent with the EMI hypothesis, they find inequalities in the placement of students in their first preference of programme/institution and in the access to more selective institutions and study programmes. Two other models were used at programme/institution level to identify the main characteristics of the programmes that explain the minimum admission GPA. They conclude that students with lower GPA, with a lower social status, will only get a place in the more selective programmes or institutions when the needs of the socially advantaged students are fully satisfied, or when these latter have secured for themselves both quantitatively and qualitatively better outcomes, which is consistent with the MMI and EMI hypotheses.

In Chap. 9, Per Olaf Aamodt presents the case of four Nordic countries—Denmark, Finland, Norway and Sweden—all based on the social democratic model characterised by redistribution policies and universalistic contribution mechanisms, high living conditions and high scores in international surveys of life satisfaction (Ramstedt, 2009) and small income differences measured by Gini-coefficients (OECD, 2016), thus creating a stark contrast with England where developments were based on neoliberal policies and marketisation of higher education.

Although there are differences among the four countries, both regarding higher education systems and education policies, which resulted in differences both in levels of inequality and in changes over time, there are however a significant number of similarities. There are no tuition fees in higher education and the introduction of tuition fees is not part of the political agenda, as fees are seen as unfair, leading to social differences in higher education attendance. Nordic countries established state-run student support systems before 1950, which make students independent of family income and financial support is given directly to the student on a universal basis (not means tested). There are less differences in prestige between higher education institutions in the Nordic countries than in other countries and, due to the relatively small income differences of the population, the relative value of higher education in terms of economic outcomes is lower than in many other countries. Aamodt quotes Erikson and Jonsson to present additional reasons for implementing policies aiming at eliminating inequalities in educational enrolments (Erikson & Jonsson, 1996).

In his chapter Aamodt argues that enrolment patterns in the Nordic countries are clearly consistent with the MMI and the EMI hypothesis. The expansion of the higher education sector was accomplished by creating a number of institutions different from universities but also less prestigious, which has led to increasing stratification within tertiary higher education. Unfortunately, and consistently with the EMI hypothesis, lower-tier opportunities are preferentially taken by students from deprived backgrounds who otherwise would not have access to tertiary education as argued in Amaral’s chapter (Koucký et al., 2010; Shavit et al., 2007). These lower-tier institutions have a more democratic enrolment than universities (Aamodt, this book) which have kept an enrolment clearly favouring students from the upper classes. As argued by Aamodt (this book), the strong social differences in enrolment to elite programmes such as law, medicine, dentistry, veterinary medicine, graduate engineering, business administration and architecture have continued.

Therefore, in the Nordic countries, despite some progress made to eliminate access inequities there is still a considerable inequality in access to higher education. Aamodt resorts to Boudon (1974) to explain the persistence of these inequities. Boudon introduced the concepts ‘primary effects’ referring to the relation between students’ ability and parental social status and ‘secondary effects’ to account for the fact that students from different social classes make different choices along their educational paths independently of their academic success (Aamodt, 1982; Erikson & Jonsson, 1996; Jackson et al., 2007). The preference of students from deprived backgrounds for vocational tracks in opposition to academic tracks, as reported in the Nordic countries and Portugal (this book) is an example of secondary effects. The use of criteria based on academic performance to determine transitions between levels in educational systems tends to increase these effects (OECD, 2008). Data collected by Hansen (2019) for Norway provide a good illustration of secondary effects by relating social class with grades, decisions to continue to upper higher education and the choice of vocational paths.

At last, Chap. 10 presents the conclusions, which are divided in three parts: a summary of the policies adopted to date in order to ensure equity in access to and success in higher education, a reflection on the inequalities that persist despite the adoption of these policies and a discussion and suggestions of possible ways forward.